Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18;
Message-ID: <c055e4e8-fd11-49e9-c7a9-bf2f1ec9de36@quicinc.com>
Date:   Wed, 9 Feb 2022 14:11:13 +0530
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:91.0) Gecko/20100101
 Thunderbird/91.5.1
Subject: Re: [PATCH v2] regmap-irq: Use regmap_irq_update_bits instead of
 regmap_write
Content-Language: en-US
To:     Marek Szyprowski <m.szyprowski@samsung.com>,
        Mark Brown <broonie@kernel.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        "Rafael J. Wysocki" <rafael@kernel.org>,
        <linux-kernel@vger.kernel.org>,
        Krzysztof Kozlowski <krzk@kernel.org>,
        'Linux Samsung SOC' <linux-samsung-soc@vger.kernel.org>
References: <20220119142953.1804-1-quic_pkumpatl@quicinc.com>
 <CGME20220208122955eucas1p2d4e32f51224242e9ebd0bce58b9c04ca@eucas1p2.samsung.com>
 <39f1b598-58ca-1e3d-3065-8dd692ee7c9f@samsung.com>
From:   Prasad Kumpatla <quic_pkumpatl@quicinc.com>
In-Reply-To: <39f1b598-58ca-1e3d-3065-8dd692ee7c9f@samsung.com>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 2/8/2022 5:59 PM, Marek Szyprowski wrote:

> There is something wrong with this patch. Since it landed in linux-next
> (20220204) I get an interrupt storm on two of my test devices:
>
> 1. ARM 32bit Exynos4412-based Trats2 ("wm8994-codec wm8994-codec: FIFO
> error" message)
>
> 2. ARM 64bit Exynos5433-based TM2e ("arizona spi1.0: Mixer dropped
> sample" message)
>
> Definitely the interrupts are not acknowledged properly. Once I find
> some spare time, I will check it further which regmap configuration
> triggers the issue, but it is definitely related to this patch.
> Reverting it on top of current linux-next fixes the issue.

Change is needed to handle the interrupt ack properly to clear the ack.

I observed that the regmap_irq_update_bits() writes the register only if 
it finds a difference b/w existing reg value to update reg value.

This may be causing the interrupt storm issue mentioned above.

Setting the mask_writeonly flag to 1, to force write the register may 
resolve the interrupt storm issue.

--Prasad